C:
Fo
od
Chemistr
y
Antioxidative Properties and Ability of Phenolic
Compounds of Myrtus communis Leaves to
Counteract In Vitro LDL and Phospholipid
Aqueous Dispersion Oxidation
Sofiane Dairi, Khodir Madani, Manar Aoun, Jos´ephine Lai Kee Him, Patrick Bron, C´eline Lauret, Jean-Paul Cristol, and Marie-Annette Carbonneau
Abstract: Antioxidant activities of Myrtus communis leaf phenolic compounds (McPCs) were investigated on 2,2 -9-azino-bis-3-ethylbenzothiazoline-6-sulfonic acid (ABTS+•) and oxygen radical absorbance capacity (ORAC) tests or on oxidation of biological models, human low-density lipoprotein (LDL) and phospholipid aqueous dispersion (l- α-phosphatidylcholine stabilized by bile salts). Two extraction techniques, microwave-assisted (MAE) and conventional (CE), were used to isolate McPCs, producing similar results of phenolic compound content. ABTS+• assay showed clearly that myrtle extracts exhibited a stronger scavenging effect than butylated hydroxyanisole andα-tocopherol, with a slight advantage for myrtle CE extract. In ORAC assay, the both McPC extracts were similarly less effective than the pure compounds as caffeic acid and myricitrin (myricetin 3-O-rhamnoside) but stronger than butylated hydroxytoluene. Moreover, myrtle CE and MAE extracts, and myricitrin were able to inhibit similarly the production of conjugated dienes and to prolong the lag phase (Tlag) during Cu2+-induced LDL oxidation with a dose-response effect. The cryo-electron microscopy observations on studied phospholipid dispersion stabilized by bile salts (BS) revealed the presence of bilayer vesicles and micelles. In 2,2-azobis (2-amidinopropane) hydrochloride–induced phospholipid/BS oxidation, myrtle CE and MAE extracts gave similar effects toα-tocopherol and caffeic acid but myricitrin showed a higher protective effect than myrtle extracts. We showed also that no synergic or additive effect betweenα-tocopherol and myrtle extracts or caffeic acid inα-tocopherol–enriched phospholipid/BS dispersion, but myricitrin showed an additive effect and thus promoted the total antioxidant activity. These data showed that myrtle extract could be used as potential natural antioxidants, food stabilizers, or natural health products.
Keywords: ABTS+• test, AAPH-mediated phospholipid dispersion oxidation, low-density lipoprotein Cu2+-mediated
oxidation, Myrtus communis, ORAC test
Practical Application: We show that microwave-assisted extraction could be an alternative method for plant phenolic compound recovery allowing important gain in time extraction. We report inhibition of low-density lipoprotein oxidation in vitro initiated by Cu2+ ions. We report that myrtle extract may be a source of natural antioxidants to counteract phospholipid peroxidation as well asα-tocopherol.
Introduction
Lipid oxidation is a complex phenomenon induced by oxygen in the presence of initiators such as heat, free radicals, photosen-sitizing pigments, and metal ions (Laguerre and others 2007) and plays an important role in many pathological processes such as atherosclerosis, cancers, Alzheimer, and Parkinson diseases. Ox-idative modification of plasma low-density lipoproteins (LDL) is
MS 20131712 Submitted 11/19/2013, Accepted 3/29/2014. Authors Dairi and Madani are with Faculty of Nature and Life Sciences, 3BS Laboratory A. Mira Univ., Bejaia 06000, Algeria. Authors Dairi, Aoun, Lauret, Cristol, and Carbonneau are with UMR 204 NUTRIPASS, - Univ., Inst. of Clinical Research – -641, Av. Doyen Gaston Giraud, 34093 Montpellier Cedex 5, France. Authors Him and Bron are with Structural Biochemistry Center, UMR 5048, 29, rue de Navacelles, 34090 Montpellier, France. Author Cristol is with Dept. of Biochemistry, Lapeyronie hospital, CHRU 371, Av. Doyen Gaston Giraud 34295 Montpellier Cedex 5, France. Direct inquiries to author Dairi (E-mail: sofianedairi@yahoo.fr).
believed to have a crucial role in the pathogenesis of atherosclero-sis (Beung 2000). In addition, oxidative stress has been recognized as a main component of post-prandial dysmetabolism (Heine and Dekker 2002; Bloomer and others 2010), suggesting that lipid ox-idation could occur during digestion process. In fact, the work of Kenmogne-Domguia and others (2014) showed that during the in vitro digestion, emulsified lipids oxidized during the gastric step, in the presence of a low pH gastric medium, but also during the in-testinal phase (presence of oxygen and oxidant substances) and that high quantities of oxidation products as malonaldehyde (MDA) and 4-hydroxy-2-hexenal (4HHE) were formed. Indeed, these oxidation products may be absorbed in the small intestine (Singh and others 2009), which may increase the oxidative stress in the plasma after their absorption. Thus, the inhibition of lipid peroxi-dation during digestion is an important role of radical-scavenging dietary antioxidants. It is now recognized that diets rich in antiox-idants derived from fruits and vegetables are associated with lower risks of cardiovascular diseases and cancers (Ishimoto and others
C:
Fo
od
Chemistr
y
2012). Thus, interest in naturally occurring antioxidants has con-siderably increased for use in food, cosmetic, and pharmaceutical products (Djeridane and others 2006). Many medicinal plants con-tain large amounts of antioxidants (phenolic acids and flavonoids), which have gained much attention as a potential source of natu-ral antioxidant phenolic compounds for improving the quality of healthcare.
Myrtus communis L. is one of the important aromatic and medic-inal species from the Myrtaceae family (Wannes and others 2010) and is typical of the Mediterranean maquis, which grows sponta-neously in many countries (Chryssavgi and others 2008). Myrtle extracts have been reported to possess antiinflammatory, antimu-tagenic, antigenotoxic, and antihyperglycaemic effects. Recent re-ports showed also that myrtle leave extracts have strongest antiox-idative activity and highest phenolic content compared with other myrtle plant parts (Chryssavgi and others 2008; Wannes and others 2010). The main class of M. communis leaf phenolic compounds was hydrolysable tannins, but the presence of phenolic acids (caf-feic acid) and flavonoids (myricitrin, a myricetin 3-O-rhamnoside) were also shown (Romani and others 1999; Wannes and others 2010). These compounds showed strong scavenging activities in many studies and can be health-promoting compounds.
To isolate antioxidants from plants, different extraction meth-ods can be used. Recently, a microwave-assisted extraction (MAE) method became an alternative extraction technique, which proved to be considerably more effective and economical. It provides higher recoveries, requires considerable less time and smaller sol-vent consumption compared to consol-ventional extraction (CE) method.
The objectives of this study were to investigate for the first time the effects of antioxidant phenolic compounds from Alge-rian M. communis leaves extracted by both methods, MAE and CE methods. For this, different antioxidant activity tests were used to evaluate the scavenging capacity of myrtle extracts. In fact, in addition to the chemical analyses (ABTS+• and ORAC), which are generally used to measure the capacity of a molecule to reduce a stable artificial free radical (by hydrogen or electron transfer), we were interested in the direct evaluation of the antioxidant ac-tivity of myrtle extracts in lipid system models oxidized in vitro: human LDL Cu+2-oxidation and 2.2-azo-bis-2-amidinopropane hydrochloride (AAPH)-induced l-α phosphatidylcholine aqueous dispersion oxidation. The last model assayed was used to simulate the duodenal (pH 6.5) and intestinal conditions (the pH was ad-justed at 7.4: data not shown) of lipid peroxidation may occurring in small intestine during lipid digestion. For this, we used bile salts as physiological detergent substances to stabilize the phospholipid aqueous dispersion, and AAPH as generator of free radicals to initialize the peroxidation. The cryo electronic microscopy was applied to visualize structure and size difference of lipid formed at pH 6.5 and 7.4 (data not shown) in the system used.
Materials and Methods
Chemicals and standards
All chemicals used were analytical grade. Egg yolk phosphatidyl-choline (EYPC), CuCl2, butylated hydroxytoluene (BHT), buty-lated hydroxyanisole (BHA), and 2N Folin-Ciocalteu reagent were purchased from Sigma Aldrich Chemical Co. (Saint Quentin Fallavier, France). Gallic acid and 2,7-dichloro-fluorescein were obtained from Merck (Darmstadt, Germany). Methanol used for chromatography was High Performance Liquid Chromatography (HPLC)-grade supplied by Merck. Ethanol used for preparing
standard solutions was from Prolabo (Paris, France). Chloroform was from Prolabo. AAPH was from Biovalley (Conches, France). Myricitrin was purchased from Roth Sochiel EURL (Lauter-bourg, France).
Plant material preparation
M. communis leaves were collected in the region of Bejaia (Algeria), and dried in an oven at 40 °C until constant weight, then crushed and sieved to have a size less than 125μm. The samples were stored in the dark at room temperature.
Extraction procedure
MAE method. A domestic microwave oven (NN-S674MF, LG, Japan, 32 l, 1000 W; variable in 100 W increments, 2.45 GHz) modified in our laboratory was used for extraction of myrtle phe-nolic compounds (McPC). A preliminary study was conducted to determine the effects of solvent type (water, ethanol/water [50/50, v/v], acetone/water [50/50, v/v], and of irradiation time [30 to 120 s]). Irradiation power at 700 W and 1/20 m/v ratio (1 g of leaves per 20 mL of solvent), respectively, were fixed. It was found that ethanol/water (50/50, v/v) or acetone/water (50/50, v/v), and 60 s of irradiation time provided the maximal recovery of McPCs (data not shown). In our study, the solvent used for extraction was chosen as ethanol/water (50/50, v/v) because it is less dangerous than acetone. The leaf suspension was irradiated by microwaves according to the following cycle: 45 s power-on (to keep temperature not rising above 80 °C), 10 s power-off (for cooling), and again 15 s power-on. The sample was filtered with a sintered glass at 0.45 μm using a vacuum pump. Two other additional extractions were carried out for recovering the totality of McPCs. Then, volume was adjusted to 50 mL. The obtained extract was stored at 4 °C until use.
CE method. The optimal parameters obtained with the mi-crowave extraction will be applied in this case: 20 mL of ethanol/water (50/50, v/v) were added to 1 g of powder and let macerate during 60 min with magnetic agitation. After that, the process was the same as for MAE method.
Quantification of McPCs
Determination of total phenolic content. Total McPC concentration was estimated by Folin–Ciocalteu’s assay, with ab-sorbance monitored at 760 nm (Mond´e and others 2011). The spectrophotometric measurement was repeated 3 times for each extract and the average data was interpolated in a gallic acid cali-bration curve and expressed on a dry weight basis as mg of gallic acid equivalents per g of dry weight sample (mg GAE g−1DW).
Determination of flavonoid content. The total flavonoids were measured by a colorimetric method as described previously (Quettier-Deleu and others 2000). Each analysis was repeated 3 times. The flavonoid concentration was expressed as milligrams of rutin equivalents per g of dry weight sample (mg RE g−1DW). Measurement of antioxidant activities
2,2-9-Azino-bis-3-ethylbenzothiazoline-6-sulfonic acid (ABTS+•) radical cation decolorization assay. The
spec-trophotometric analysis of ABTS+• scavenging activity was de-termined according to a method (Re and others 1999) based on the ability of antioxidants to quench the long-lived cation radical, in comparison to that of BHA a synthetic antioxidant, or α-tocopherol. The ABTS+• was produced by reacting 7 mM ABTS in H2O with 2.45 mM potassium persulfate, stored in the dark at room temperature. Before use, the ABTS+• solution was diluted
C:
Fo
od
Chemistr
y
to get an absorbance of 0.700± 0.020 at 734 nm with ethanol. Then, 1 mL of ABTS+• solution was added to 1 mL of McPCs or standard ethanolic solutions at different concentrations. Inhibition was evaluated as percentage reduction of absorbance for each sam-ple concentration expressed as mg GAE L−1for McPCs or as mg L−1for pure standards. The IC50 (concentration providing 50% inhibition) was calculated.
Oxygen radical absorbance capacity (ORAC) as-say. ORAC values were measured with fluorescence spec-trometer (Victor2 Wallac–Perkin-Elmer) by inhibition of 2,7dichlorofluorescein (DCF; Ishimoto and others 2012) with slight modifications. Briefly, all samples and reagents were dis-solved in 10 mmol phosphate L−1/150 mmol NaCl L−1 buffer (PBS) at pH 7.4. 50μL test samples or 50 μL Trolox solutions (0 to 20μmol L−1), 100μL DCF solution (50 nmol L−1), and 100μL AAPH solution (20 mmol L−1) were added to the wells of a 96-well plate. The fluorescence was recorded every 1 min for 90 min at 485-nm excitation and 535-nm emission wavelengths. A calibration curve of ORAC levels was obtained by plotting the period of time needed to obtain 50% fluorescence decay versus the Trolox concentrations. ORAC levels were expressed as mole of Trolox equivalent (TE) per mole of antioxidant (pure com-pounds) or mole of GAE (extracts). Gallic and caffeic acids were used as controls. Their respective ORAC values of 1.3± 0.4 and 4.1± 0.5 mol TE mol−1(means± SD) were found close to those previously reported (Ishimoto and others 2012).
LDL isolation and evaluation of PC extract effect on LDL oxidation mediated by Cu2+ ions. LDL was isolated from fresh plasma of healthy human subjects, obtained from the “French Blood Establishment,” in accordance with its ethical rules, and oxidizability was monitored at 234 nm for 5μmol L−1 Cu2+oxidation as previously indicated (Mond´e and others 2011). Briefly, isolated LDL was diluted to 1μmol apoB L−1, added with the various McPC concentrations to be tested, and then 10-fold diluted in oxygenated PBS at pH 7.4. For rendering the antiox-idant abilities, we used the notion of specific antioxantiox-idant activity (SAA), which was calculated as the slope of the linear relationship obtained between relative lag time (rTlag), and concentrations of the different tested compounds (Mond´e and others 2011). Relative Tlag was defined as [Tlag+/Tlag−]× 100, with+and− denot-ing LDL with and without antioxidants, and Tlag was defined as the time corresponding to the end of the 1st kinetic phase dur-ing, which optical density (OD) do not or only slowly increase. An increased protection ratio was calculated using the follow-ing formula: protection ratio (%)= [(Tlag+– Tlag−)/Tlag−]× 100, with+ and − denoting LDL with and without tested an-tioxidants. Finally, the CDmaxvalue was calculated as ODmax/ε, where ODmaxcorresponded to maximal oxidized product accu-mulation that was determined graphically by drawing a tangential line at the highest point corresponding to the end of the oxida-tion propagaoxida-tion phase, andε is the specific absorption coefficient of CDs. The rate of oxidation propagation (Rp) was then ex-pressed as mol CDmax mol apoB−1 min−1. An inhibitory ratio was finally calculated using the following formula: inhibitory ratio (%)= [Rp−– Rp+/Rp−]× 100.
Preparation of phospholipid aqueous dispersions with EYPC and bile salts and evaluation of McPC extract effects on oxidation mediated by AAPH. An appropriate quantity of EYPC was dissolved in chloroform. An aliquot of EYPC solu-tion was dried under nitrogen, and bile salts (50%-cholic and 50 %-desoxycholic sodium salts) were added. For samples enriched withα-tocopherol, α-tocopherol was added to the EYPC
solu-tion before drying. The aqueous dispersion was prepared by adding 10 mL of PBS at pH 6.5, to make stock dispersions of 2.96 and 3.38 g L−1 of EYPC and bile salt concentrations, respectively. Then, sonication was carried out, under a stream of N2and in an ice bath, with 3 to 5 30-s irradiation cycles with a delay time of 10 s, until complete dispersion of EYPC. The aqueous dispersion was incubated at 37 °C during 30 min, and filtered to 0.2μm. Samples were sealed under nitrogen and stored at−80 °C.
The oxidative stability of the emulsion was determined by mon-itoring the formation of conjugated dienes (CDs) with a UV-Visible thermostatic Spectrometer (Uvikon-XL; Bio-TeK Instru-ments) at 245 nm and 37 °C for 150 min with measurement intervals of 1.5 min. For this, to 1 mL of phospholipid aqueous dispersion were added 0.5 mL of AAPH (as oxidant agent) with a final concentration of 5 mmol L−1, and 0.5 mL PBS contain-ing different tested concentrations of McPCs or pure standards. Oxidation results were calculated as the ratio of AUC of tested antioxidants to that of control, and were expressed as percentages (Lorrain and others 2010). The AUC value was calculated by the integration of the area under the curve (AUC) obtained by CD concentrations versus time (until 150 min).
Extraction of lipids, determination of fatty acid (FA)
composition, andα-tocopherol content
Total lipids were extracted from aqueous dispersion according to a slight modification of Folch’s technique (Folch and others 1957). Briefly, a mixture containing 250 μL of dispersion and 750μL of 9 g NaCl L−1was homogenized with 4 mL of chlo-roform/methanol (2/1, v/v) containing 50 mg L−1 of BHT as an antioxidant. This homogenate was centrifuged to recover total lipids in the organic phase, which were then washed using 9 g NaCl L−1and recentrifuged. The total recovered organic phase was evaporated under N2and the residue was taken again by 2 mL of a chloroform/methanol (2/1, v/v) mixture, and then stored at −20 °C until further analyses. Preparation of the methylated FAs and gas chromatography analyses were carried out according to our usual technique (Aoun and others 2011).
α-Tocopherol was determined after extraction as previously de-scribed (Mond´e 2011) with slight modifications. Briefly, 1 mL of aqueous dispersion was extracted by ethanol/hexane (1/1, v/v) containingδ-tocopherol as an internal standard. Extracted prod-ucts were measured after HPLC separation with spectrophotomet-ric detection (at 292 nm), on a LichrocartR 125-4 (5μm-particle
size) column (Merck, France), using a water/methanol mixture (3/97, v/v) mobile phase at 0.8 mL min−1flow rate for 12 min. Cryo-transmission electronic microscopy
Sample freezing was performed using a semiautomatic plunge freezing instrument (cryoplunge CP3; Gatan Inc.). Briefly, 3 mi-croliters of phospholipid aqueous dispersion at 3 g/L were applied to glow discharged quantifoil R 2/2 grids (Quantifoil Micro Tools GmbH, Jena, Germany), blotted for 2 s and then flash frozen in liq-uid ethane. Cryo-transmission electron microscopy (cryo-TEM) observations were carried out on a JEOL 2200FS FEG operating at 200 kV under low-dose conditions (total dose of 20 electrons/ ˚A2) in the zero energy loss mode with a slit width of 20 eV. Images were recorded with defocus ranging from 1.5 to 2μm.
Statistical analysis
Analyses were carried out in 3 times or more and results were reported as mean values ± standard deviation (SD). Data were compared on the basis of the mean values. Differences among
C:
Fo
od
Chemistr
y
means of variety groups were tested using a Tukey–Kramer HSD (logiciel JMP version 7.0) with a significance level of 0.05.
Results and Discussion
Quantification of M. communis phenolic compounds Total McPC content was determined by Folin–Ciocalteu’s assay. The both MAE and CE methods gave almost very close levels of phenolic compounds (176.4± 6.8 and 181.0 ± 6.3 mg GAE g−1 DW, respectively), but with a slight advantage for the CE method
(P< 0.05). Results of flavonoid content showed that MAE method was more effective than CE one (25.6± 0.9 and 22.7 ± 0.6 mg RE g−1 DW, respectively).
According to our results, MAE method showed to be an al-ternative to the conventional one, as previously shown for many plants (Proestos and Komaitis 2008). Microwave energy offers a rapid transfer of energy to the extraction solvent and raw plant ma-terials. This extracting method also results in rupture of the plant cells and quickly release of intracellular products into the solvent. McPC contents obtained with the 2 extraction procedures were
0 20 40 60 80 100 0 0.5 1 1.5 2 Inhibition % Concentration mg GAE L-1 M.communis CE M.communis MAE BHA α tocopherol
Figure 1–Scavenging activity of M. communis extracts on ABTS radical. Results are expressed as means of 3 experiments± SD (n= 3). 0 0.2 0.4 0.6 0.8 1 1.2 0 20 40 60 80 100 Relative fluor escence intensity Time (min) Control M.communis CE M.communis MAE Myricitrin BHT Caffeic acid
Figure 2–Fluorescence decay curve of fluorescein during the ORAC assay in the presence of various phenolic compounds (1 μmol L−1). The results are the mean of 3 experiments (n= 3).
C:
Fo
od
Chemistr
y
found to be 2 times lower than that found in methanolic extract of Greece myrtle leaves (Chryssavgi and others 2008).
Antioxidant activities of McPCs
The ABTS+• assay allowed to determine the electron-donating capacity of antioxidant compounds and showed that myrtle ex-tract obtained by CE method and BHA standard exhibited higher antioxidant capacity (Figure 1) with an IC50of 0.68± 0.10 mg GAE L−1and 0.89± 0.07 mg L−1; respectively, followed by MAE method with an IC50of 1.05± 0.02 mg GAE L−1, and finally the lowest antioxidant capacity was obtained byα-tocopherol standard with an IC50= 1.49 ± 0.11 mg L−1(P< 0.05) compared to the 3 other compounds. Hence, when data are expressed in terms of dry weight (instead of GAE), the myrtle (CE and MAE) extracts with an IC50value of 3.79± 0.13 and 5.95 ± 0.22 mg DW L−1 re-spectively, became less effective than the 2 pure compounds tested. This supposes that in myrtle extract, there are other compounds devoid of antioxidant activity that led to underestimate the antiox-idant activity of the extract when the result was expressed on dry weight basis. Our results were in accordance with a previous work on Moroccan myrtle leaves showing the ability of myrtle extract to scavenge ABTS+• radical (Amensour and others 2010). It was also found that MAE method presented a less scavenging activity than CE one. These results lead to the hypothesis that microwave-induced heating of extraction solvent and this might cause a ther-mal degradation of some antioxidant compounds present in the extract during the irradiation phase, as shown previously (Liazid and others 2007). These authors concluded that all the compounds studied are stable up to 100 °C, whereas at 125 °C there is sig-nificant degradation of epicatechin, resveratrol, and myricetin. It has also been found that phenolic compounds having a greater number of hydroxyl substituents are more easily degraded under the extraction conditions, which let suppose particularly a high potential degradation of myricetin derivatives possibly present in the extract. Moreover, we can also hypothesize that in both ex-tracts, there is a difference of antioxidant substance composition due to the both used different extraction procedures.
The 2nd chemical assay used was ORAC test, which could be considered as a preferable method because of its biological relevance to the in vivo antioxidant efficacy (Oueslati and others 2012) and where the operating mechanism is H-atom transfer reactions from the phenols to AAPH-derived peroxyl radicals. Figure 2 showed that myrtle extracts obtained by MAE and CE methods showed similar results with an ORAC value of 1.59± 0.07 and 1.57 ± 0.13 mol TE mol GAE−1, respectively, and were less effective (P < 0.05) than pure standards as myricitrin and caffeic acid, with ORAC values of 6.3± 0.4 and 4.1 ± 0.5 mol TE mol−1. However, they were more effective than BHT as synthetic antioxidant with an ORAC value of 0.7± 0.1 mol TE mol−1.
These values confirm that an increase in the number of hydroxyl groups, particularly in the B-ring of flavonols (myricetin) is gen-erally correlated with an increase in antioxidant activity (Cao and others 1997). Moreover, myricitrin has a slightly increased ORAC value than myricetin, its aglycon form (4.9± 0.2 mol TE mol−1, as previously found by Ishimoto and others 2012), possibly via its rhamnoside moiety. A recent work (Oueslati and others 2012) showed also that pure antioxidant as caffeic acid or quercetin have a stronger scavenging effect than 4 acetonic extracts of Tunisian halophytes. These results may be explained by a competition for the H-donating capacity between different antioxidants present in a natural mixture.
Myrtle extracts prevent copper-induced LDL oxidation It is important to mention that plasma concentrations achieved by flavonoids are low, usually no more than 1μmol L−1(Halliwell and others 2005), a reason to choose the concentrations used under our LDL oxidization model.
Effects of McPCs (MAE and CE) were evaluated on 5μmol Cu2+L−1-mediated LDL oxidation at different concentrations (0 to 2.95μmol GAE L−1). Oxidation process was assessed by kinetic of CD formation (Figure 3) evaluated by OD at 234 nm. Myrtle extracts were able to inhibit in a dose-dependent manner LDL oxidation leading to a significant increase in lag time (Table 1).
0 0.2 0.4 0.6 0.8 1 0 50 100 150 200 250 300 Absorbance ( 234 nm) Time (min) LDL control LDL + 0.74 μmol GAE L-1 LDL + 1.48 μmol GAE L-1 LDL + 2.21 μmol GAE L-1 LDL + 2.95 μmol GAE L-1
Figure 3–Kinetic of CDs production by LDL Cu2+-mediated oxidation in the presence of McPCs (MAE method).
LDL (0.1μM) is oxidized in PBS pH 7.4 at 37 °C with 5μmol Cu2+L−1with McPCs added just before oxidation, or without antioxidant for LDL control. Concentrations of McPCs are evaluated asμmol GAE L−1and the absorbance is continuously monitored at 234 nm. CD production is evaluated as OD/ε of CDs. Kinetic of CD production with McPCs (CE method) gives similar curves.
C:
Fo
od
Chemistr
y
Table 1–Effect of polyphenol extracts of M. communis leaves obtained by MAE or CE methods on the kinetic parameters of LDL oxidation mediated by Cu2+ions.
RP(CD-mol CDmax(CD-mol
Tlag (min) apoB-mol−1min−1) apoB-mol−1)
concentration
(μmol GAE L−1) MAE CE MAE CE MAE CE
0 41.8± 1.7a 48.9± 1.8a 8.4± 0.2a 7.3± 0.3a 277.7± 7.5a 246.22 ± 9.4a 0.74 60.6± 1.1b 80.0± 3.6b 8.1± 0.2a 6.7± 0.2ab 276.8± 9.0a 247.7± 19.5a 1.48 92.9± 14.3c 142.3± 2.8c 8.1± 0.8a 6.7± 0.4ab 276.4± 9.9a 240.4± 10.8a 2.21 130.6± 13.9d 182.2± 21.2d 8.0± 0.3a 5.9± 0.7ab 270.3± 13.6a 234.0± 5.1a 2.95 180.4± 4.5e 245.9± 12e 7.0± 0.3b 5.8± 0.3b 255.5± 7.0a 225.8.0± 2.6a
Assays are performed in triplicate (n= 3) and data are expressed as mean ± standard deviation. Means followed by different letters in the same column are significantly different (P< 0.05). Kinetic parameters of LDL oxidation are: Tlag, lag time of oxidation curve; Rp, oxidation propagation rate; CDmax, maximal oxidized product accumulation evaluated
as ODmax/ε of CDs.
Moreover, at higher concentration (2.95μmol GAE L−1), McPCs allowed to produce a significant (P< 0.05) increased protection ratio of 332.5 ± 28.1% and 402.9 ± 8.8% for MAE and CE methods, respectively. Moreover, the oxidation propagation rate (Rp) decreased (P< 0.05), giving an inhibitory ratio of 16.4 ± 5.5% and 20.7± 3.4% for MAE and CE methods, respectively, whereas the formation of oxidation products evaluated as CDmax remained constant. Finally, SAA values for McPC extracts were also evaluated, and the results were expressed as μmol GAE−1 L (Figure 4) in comparison with pure standards, for which SAA values were expressed asμmol−1L. Higher SAA value indicated higher antioxidant activity. Myrtle extracts (MAE and CE) and myricitrin exhibited SAA values without significant difference (130 ± 5.5, 102.0 ± 7.3 μmol GAE−1 L, and 110.2 ± 17.3 μmol−1L, respectively) on LDL containing naturalα-tocopherol level (8.3± 0.6 mol α-tocopherol mol apoB−1). However, these 3 SAA values were significantly higher (P< 0.01) than that of caffeic acid orα-tocopherol (54.0 ± 8.6 or 7.2 ± 0.3 μmol−1L, respectively), values obtained on LDL exhibiting an exceptionally high α-tocopherol content (12.7 ± 1.1 mol α-tocopherol mol apoB−1). The very low level of protection afforded by an external addition of α-tocopherol could be explained by its lack in the catechol structure affording maximal radical stabilization.
LDL oxidation is a complex, multistep mechanism involving both lipid and protein fractions. According to the previously pro-posed mechanism (Pinchuk and Lichtenberg 2002), (1) if the an-tioxidant prolongs the lag time without affecting the maximal rate, the underlying mechanism of inhibition probably involved either quenching of free radicals or prevention of the transfer of reactive intermediates (particularly hydroperoxides) to the LDL lipid core; and (2) if the maximal rate of Cu2+peroxidation was decreased, the conclusion is that the mechanism involved either binding of copper ions or blocking of copper binding sites on the lipopro-tein surface LDL. So, McPC protective effects could be afforded by a combination of these both mechanisms. Moreover, when myricitrin was used under the same oxidative condition, the Rp of oxidation remained constant. Myricetin, its aglycone form, was an efficient antioxidant for protecting LDL against Cu2+-mediated oxidation and may act by chelating copper ions due to the pres-ence of catechol group (3and 4OH) and/or donating hydrogen atom to generated peroxyl radicals (Vaya and others 2003). Caffeic acid, one of the most abundant phenolic acids in myrtle leaves (Romani and others 1999), inhibited strongly LDL modification with a dose-dependent effect (Mond´e and others 2011). It was also found that McPCs of the CE method had a slightly higher SAA than that of MAE method and myricitrin but without a
a a a b c 0 20 40 60 80 100 120 140 160
M.com CE M.com MAE Myricitrin Caffeic acid α tocopherol
SAA (
μ
mol GAE
-1 L)
Figure 4–SAAs values of M. communis extracts, myricitrin, caffeic acid, andα-tocopherol for Cu2+-induced LDL oxidation assay.
SAA= specific antioxidant activity, defined as the slope of the relationship between relative Tlag and concentrations of tested compounds. Results are expressed as means of 3 experiments ± SD. Different letters indicate that samples are significantly different (P< 0.05).
C:
Fo
od
Chemistr
y
significant difference (Figure 4). This result did not correlate with chemical tests, particularly the ABTS+• assay giving statistically different values for McPCs issued from MAE and CE methods. This observation was also different with the ORAC assay, which showed that pure compound (myricitrin) had largely more in-hibitory effect than the both myrtle extracts. Our results upon LDL protective effect may be interpreted by possible competition of the different compounds present in myrtle extracts. As previ-ously shown (Meyer and others 1998), different combinations of phenolic compounds may have additive effects, except for combi-nations including ellagic acid and catechin, present together in M. communis extracts (Romani and others 1999), where ellagic acid could exert a significant antagonistic effect.
Myrtle extracts prevent AAPH-induced oxidation of phospholipid aqueous dispersion
The effect of McPCs (MAE and CE) was also evaluated on AAPH oxidation of EYPC/bile salt aqueous dispersion with the ratio of 0.8 (EYPC/BS) in PBS pH 6.5. Gas chromatography analysis of FAs showed that this aqueous dispersion contained an equivalent composition between saturated and unsaturated FAs. Major saturated FAs were palmitic (29.5± 4.0%) and stearic (16.9
± 2.1%) acids. Major unsaturated FAs were in a decreasing order, oleic (25.2± 2.9%), linoleic (14.1 ± 1.4%), arachidonic (4.6 ± 0.5%), docosahexaenoic (3.9± 1.1%), and linolenic (1.7 ± 0.1%) acids.
The lipid structures in the phospholipid aqueous dispersion studied were characterized by cryo-transmission electron mi-croscopy (cryo-TEM), a well-adapted technique for observing structures in complex fluids such as mixed micellar dispersions, to identify the lipid structures formed by addition of the surfactant bile salts to phosphatidylcholine dispersion (Almgren and others 2000). All images revealed the presence of 2 different types of lipid structures, clearly indicating heterogeneity of the dispersion (Figure 5A and 5B). Unilamellar vesicles of about 200 nm in di-ameter are predominantly found in our lipid dispersion. These structures, hollowed and delineated by a bilayer are consistent with liposomal forms. It should be mentioned that some larger vesicles can also be present, as showed by a vesicle of 346 nm in diameter in Figure 5C and that membrane of some rare vesi-cles can locally be disrupted (white arrow). Small dense partivesi-cles, ranging from 15 to 40 nm in diameter, are also observed in the dispersion, suggesting the presence of mixed micelles (black ar-rows, Figure 5). Our results are consistent with a previous work
Figure 5–Cryo-electron microscopy images of phospholipid aqueous dispersion formed by lecithin (EYPC) and bile salts at pH 6.5. (A and B) Representative frozen-hydrated lipid vesicles observed in phospholipid dispersion, with diameter ranging from 200 to 250 nm. Small dense micelles can also be visualized as shown by black arrows. C, lipid vesicle having diameter larger than 300 nm can occasionally be observed as well as breach in liposome lipid bilayer (black arrow). Scale bars, 100 nm.
C:
Fo
od
Chemistr
y
(Hildebrand and others 2004) that studied the solubilization of 1, 2-dipalmitoyl-sn-glycero-3-phospha-tidylcholine liposomes by bile salts, and demonstrated that this surfactant agent form large mixed vesicles and small mixed micelles. In our study, the presence of a large mixed vesicles (diameter> 200 nm) may be explained by the coalescence phenomena corresponding to fusion of vesicles. This is supported by the presence of punctual membrane breaking observed in our dispersion images (Figure 5C).
During aqueous dispersion AAPH oxidation at pH 6.5 (Figure 6), McPCs and myricitrin showed that they were able to minimize the formation of CDs evaluated from ODmax, with a dose-dependent effect. In fact, when McPC (MAE and CE) concentrations increased, AUC values decreased significantly (P< 0.05) with a linear relationship giving equivalent IC50values of 16.1 and 15.5μmol GAE L−1, respectively). Myricitrin and caffeic acid were more than and as efficient as myrtle extracts, respectively. However, these both standard compounds gave a 2nd degree polynomial relationship with an IC50 value of 8.1μmol L−1for myricitrin and only a maximum inhibition value of 57.7% obtained for a concentration of 17.7μmol L−1for caffeic acid.
This observation coincided with results found in the ORAC as-say showing that myricitrin was more effective than myrtle extracts. A previous study (Gulcin 2006) showed also that caffeic acid was an effective antioxidant on peroxidation of linoleic acid emulsion. It was important to mention also that linoleic acid was the main oxi-dizable polyunsaturated FA found in our phosphatidylcholine/bile salt dispersion. Moreover, the antioxidant activity of myricetin and myricitrin in bulk methyl linoleate system thermally oxidized at 40 °C was analyzed (Hopia and Heinonen 1999). Their results showed that the antioxidant activity difference between myricetin and its rhamnoside derivative was small but statistically signifi-cant, and they were more active than equivalent concentrations ofα-tocopherol. Under our conditions, no significant difference was obtained between myricitrin and α-tocopherol protection. Comparable results were also obtained for quercetin and rutin, its glycolysated form compared to α-tocopherol (Lorrain and oth-ers 2010) on sunflower oil-in-water emulsions, stabilized by egg yolk phospholipids and oxidized by methmyoglobin. This
sug-gested that these hydrophilic antioxidants were as efficient as the lipophilicα-tocopherol to protect sunflower oil-in-water emul-sions or mixed phospholipid/bile salt disperemul-sions against oxidation mediated by lipophilic metmyoglobin-derived or by hydrophilic AAPH-derived peroxyl radicals, respectively.
Myrtle extracts does not act synergistically with
α-tocopherol on phospholipid aqueous dispersion oxidation
In this study, we also investigated the effect of added α-tocopherol on the stability of lipid dispersions, and possible syn-ergic effects with other antioxidants, at pH 6.5. First, as shown in Figure 7A, at an overloading concentration of 10μmol L−1, α-tocopherol, caffeic acid, and myricitrin decreased the AUC value to 50.4± 7.5%, 60.3 ± 3.1%, and 44.4 ± 7.7%, respec-tively, as compared to the control without any added antioxidant. McPCs, at an equivalent concentration of 10 μmol GAE L−1, decreased the AUC value to 67.8± 2.2% and 69.0 ± 1.9% for both MAE and CE methods, respectively. These results showed thatα-tocopherol and myricitrin had an antioxidant effect signif-icantly (P< 0.05) higher than that of both McPCs. Moreover, when McPCs (MAE or CE, 10μGAE L−1) were added in addi-tion toα-tocopherol (10 μmol L−1), protection of the dispersion was more effective (ca. a factor 1.5 or 1.6) than that provided solely by McPCs. However, this gain in efficiency is lower than that could be anticipated by an additive effect. Second, it was also investigated possible synergic effects of myrtle extracts, caf-feic acid and myricitrin (Figure 7B) on α-tocopherol–enriched phospholipid aqueous dispersions. The dispersion control contains 3.7± 0.9 μmol α-tocopherol L−1, which is naturally present in the EYPC, whereasα-tocopherol–enriched dispersion contained 10.6± 0.2 μmol L−1. The results showed that there was no differ-ent protective effect ofα-tocopherol between its external loading (decreased AUC to 50.4 ± 7.5% for a total [initial + external loading] 14.3μmol α-tocopherol L−1, as shown in Figure 7A) and its incorporation in the lipid phase prior to the aqueous dis-persion formation (decreased AUC to 67.3± 10.9% for only 10.5 μmol α-tocopherol L−1, as shown in Figure 7B). Moreover, no more additive effects were obtained by external addition of McPCs
0 0.2 0.4 0.6 0.8 1 0 50 100 150 200 Absorbance (245 nm) Time (min) control
M. com 3.78 μmole GAE L-1 M. com 7.57 μmole GAE L-1 M. com 11.35 μmole GAE L-1 M. com 15.14 μmole GAE L-1
Figure 6–Kinetic oxidation of phospholipid aqueous dispersion formed by egg
phosphatidylcholine and bile salts at pH 6.5, induced by AAPH-derived radical in the presence or absence (phospholipid dispersion control) of M. communis (M.com) MAE-PCs. Phospholipid aqueous dispersion was oxidized by 5 mmol AAPH L−1with or without M.
communis PCs (phospholipid dispersion
control). Phenolic compounds were added just before oxidation. Concentrations of McPCs were evaluated asμmol GAE L−1and the absorbance was continuously monitored at 245 nm. CD production is evaluated as OD/ε of CDs. Kinetic of CD production with M.
C:
Fo
od
Chemistr
y
(MAE and CE; 10μGAE L−1), or caffeic acid (10μmol L−1) on the α-tocopherol-enriched dispersion, by comparison with the control. Only a slight significant additional protective effect (ca. a factor 1.4) was obtained for external addition myricitrin (10μmol L−1) onα-tocopherol–enriched dispersion.
This finding can be supported by a previous work (Yin and others 2012) who investigated the antioxidant interaction between green tea polyphenols (GTE) andα-tocopherol against lipid per-oxidation. A merely additive effect of α-tocopherol and GTE was observed in their phosphatidylcholine-based liposome system, when a synergistic effect was only observed in a sunflower oil-in-water emulsion, suggesting that this GTE extract is more efficient as antioxidant in lipid systems containing triglycerides afforded by sunflower oil. One another study (Peyrat-Maillard and oth-ers 2003) observed an antagonistic effect betweenα-tocopherol
and some phenolic acids (particularly rosmarinic and caffeic acids) during AAPH-oxidation of an aqueous dispersion of linoleic acid. These authors explained this antagonism by the fact that a fraction of highly active acids (rosmarinic and caffeic acids) would regen-erate the less activeα-tocopherol, besides giving hydrogen to lipid (alkyl or peroxyl) radicals. The phenol group of α-tocopherol is located at the interface of the aqueous phase (Altunkaya and others 2009). Thus, this liposoluble antioxidant can trap free rad-icals generated either by lipids at the lipid–water interface where peroxidation was predominant, or by AAPH decomposition oc-curring into the aqueous phase. Myrtle extracts, as a mixture of different antioxidants (phenolic acids and flavonoids), probably with difference in solubility and mechanism of action, could neu-tralize both AAPH-initiated aqueous radicals and lipid radicals at the interface. This may explain a similar effectiveness of myrtle
a bc bc def f ef cde ef 0 20 40 60 80 100 120 control M.com
CE M.comMAE αtoc αtoc+M. com CE αtoc+M. com MAE caffeic acid myricitrin AUC %
A
a bc bcd b bc def 0 20 40 60 80 100 120control control αtoc M.com CE M.com
MAE caffeic acid myricitrin
AUC %
B
Figure 7–Evaluation of antioxidant activities and synergic or antagonist effects of myrtle extracts and pure compounds (α-tocopherol, caffeic acid, and myricitrin) on the stability of phospholipid aqueous dispersion unenriched (control) (A) and enriched (B) with α-tocopherol (control αtoc), at equivalent concentration (10μGAE L−1for myrtlePCs, and 10μmol L−1for pure compounds). AUC% (percentage of AUCs) was the ratio between (AUC+/AUC−)× 100, where AUC+and AUC−are the AUC with and without antioxidant, respectively. Results are expressed as means of 3 experiments± SD. Different letters indicate that samples are significantly different (P< 0.05).
α-Tocopherol in unenriched (A) and enriched (B) phospholipid dispersion were 3.7± 0.9 and 10.6± 0.2 μmol L−1, respectively.
C:
Fo
od
Chemistr
y
extract and α-tocopherol, in our phospholipid aqueous disper-sion model. Myricitrin was the best active antioxidant, which may be explained by a better reactivity with AAPH-derived radicals like demonstrated above in the ORAC assay, but also its possible localization near the interface where it could interact with lipid radicals.
These different observations could also explain the difference concerning McPC protective effects observed between our 2 bi-ological models, LDL containing a large variety of lipids (choles-terol, triglycerides and phospholipids + natural α-tocopherol) with an apoproteinB-containing interface and mixed lipid disper-sion containing only egg yolk phospholipids/bile salts+ natural α-tocopherol. In a previous study, we showed that in oxidized LDL model, caffeic acid or phenolic compounds from extracted red wine or oil palm ripe fruits (Mond´e and others 2011) were able to delay the in vitro LDL-α-tocopherol oxidation by metal-dependent (Cu2+) or -independent (AAPH) initiation, yielding confirmation of a reducing effect on the oxidizedα-tocopherol. This previous observation and this study underline that in the LDL model, McPCs are more efficient thanα-tocopherol and a part of them is used to regenerateα-tocopherol.
The chemical assays showed only the ability of myrtle extracts to scavenge free radicals by electron or H atom transfer, and lack a biological significance because the peroxidation occur in a com-plex system where different ways of antioxidant implication may be explored. This may explained the difference found sometimes between chemical assay results (ORAC test) and lipid model ox-idation results. The lipid systems used in simulated physiological conditions revealed different mechanism by which antioxidants may act against lipid peroxidation. The Cu+2-induced LDL per-oxidation showed a structure–activity relationship, which may ex-plain the effectiveness of myrtle extracts and myricitrin more than α-tocopherol, which is devoid of a catechol structure. Moreover, McPCs may act synergically withα-tocopherol by regenerating tocopheroxyl radical, during LDL oxidation. In dispersed lipid system, the localization of antioxidants is more important to de-termine the antioxidant activity. For this,α-tocopherol was a po-tent antioxidant as well as myrtle extracts. Thus, McPCs possess a potent chelating and free radicals scavenging activities. More-over, the method ofα-tocopherol addition to the lipid dispersion did not affect the antioxidative properties of this lipid soluble antioxidant. Gal and others (2003) who studied the palmitoylli-noleoylphosphatidylcholine (PLPC) liposome oxidation mediated by copper (Cu2+) reported that pro-oxidative effects could be observed whenα-tocopherol was added externally to PLPC li-posome whereas, whenα-tocopherol was co-sonicated with the phospholipids, it acted as an antioxidant. Thus, the types of oxi-dant used determine the outcome of the antioxioxi-dant activity assay. Different interactions between phenol-containing myrtle extract, α-tocopherol, and lipid bilayers may occur (Gutierrez and oth-ers 2003). These interactions led to an antagonist effect when these antioxidants were tested together in our conditions. For the antagonism observed with the combination of McPCs and α-tocopherol in liposome model, the potential mechanism remains unclear. According to the work of Yin and others (2012), it may be suppose that this phenomena could be attributed to: (a) the physical barriers between antioxidants and the lipid derived rad-ical; and (b) a process in which antioxidant radicals formed via the autoxidation of the less effective antioxidant oxidize the more effective antioxidant, thus preventing the latter to inhibit lipid ox-idation. Indeed, the work of Becker and others (2007) showed that the combination ofα-tocopherol and quercetin had a
syner-gistic effect only when quercetin was at high concentration and α-tocopherol at low concentration in liposomal phospholipid ox-idation. This is another aspect to consider also. The interactions responsible for this effect require further investigations with respect to their biological significance.
Conclusion
This study showed that antioxidant efficacy is dependent of the lipid substrate, and the oxidation conditions used, which may explain the difference found sometimes in antioxidant power clas-sification of the same molecule evaluated in different tests, for example the difference found for myrtle extract scavenging effect in LDL and phospholipid assay. MAE extraction method showed to be an alternative technique to the conventional one giving the same antioxidant activities. Phenolic compounds of M. communis leaves possess an effective antioxidant activity toward free radical (ABTS+• and ORAC tests), and oxidation induced by Cu2+and AAPH of biological models, LDL and phospholipid/BS aqueous dispersions, respectively. Moreover, it was established in the phos-pholipid oxidation assay, that none of the compounds, at the tested levels, exerted any antioxidant synergism withα-tocopherol. Ac-cording to all these data, McPCs could to be a good source of natural antioxidants and health promotion products. Thus, lipid oxidative stress has been recognized as a main component of post-prandial digestion process, so McPCs could exert direct in vivo beneficial antioxidant effects by protecting dietary polyunsaturated lipids in the intestinal micellar system.
Acknowledgments
The authors thank Dr. Begdouche of Botany Dept., Univ. of Bejaia (Algeria) for his identification of M. communis leaves.
Conflicts of Interest
No conflict of interest in our present study.
Authors’ Contributions
K. Madani, M. A. Carbonneau, and J. P. Cristol supervised all the study, analyzed and interpreted the data, and they corrected the manuscript. M. Aoun analyzed FAs composition of phospholipid aqueous dispersion studied and corrected the manuscript. J. Lai Kee Him and P. Bron analyzed by cryo-electron microscopy the lipid structure in the phospholipid dispersion and corrected the manuscript. C. Lauret contributed to vitamin E analyses.
References
Altunkaya A, Becker EM, G ¨okmen V, Skibsted LH. 2009. Antioxidant activity of lettuce extract (Lactuca sativa) and synergism with added phenolic antioxidants. Food Chem 115(1):163–8. Almgren M, Edwards K, Karlsson G. 2000. Cryo transmission electron microscopy of liposomes
and related structures. Colloids Surf A: Physicochem Eng Aspects 174:3–21.
Aoun M, Michel F, Fouret G, Schlernitzauer A, Ollendorff V, Wrutniak-Cabello C, Cristol JP, Carbonneau MA, Coudray C, Feillet-Coudray C. 2011. A grape polyphenol extract modulates muscle membrane fatty acid composition and lipid metabolism in fat high-sucrose diet-fed rat. Br J Nutr 107(5):647–59.
Amensour M, Sendra E, Abrinia J, Perez-Alvarez JA, Fernandez-Lopez J. 2010. Antioxidant activity and total phenolic compounds of myrtle extracts. J Food 8(2):95–101.
Becker EM, Ntouma G, Skibsted LH. 2007. Synergism and antagonism between quercetin and other chain-breaking antioxidants in lipid systems of increasing structural organization. Food Chem 103:1288–1296.
Beung HR. 2000. Low-Density Lipoprotein (LDL), Atherosclerosis and Antioxidants. Biotech-nol Bioproc Eng 5:313–9.
Bloomer RJ, Kabir MM, Marshall KE, Canale RE, Farney TM. 2010. Postprandial oxidative stress in response to dextrose and lipid meals of differing size. Lipids Health Dis 9:79–89. Cao G, Sofic E, Prior R L. 1997. Antioxidant and proxidant behavior of flavonoids:
structure-activity relation-ship. Free Radical Biol Med 22(5):749–60.
Chryssavgi G, Vassiliki P, Athanasios M, Kibouris T, Michael K. 2008. Essential oil composition of Pistacia lentiscus L. and Myrtus communis L.: Evaluation of antioxidant capacity of methanolic extracts. Food Chem 107:1120–30.
Djeridane A, Yousfi M, Nadjemi B, Boutassouna D, Stocker P, Vidal, N. 2006. Antioxidant activity of some Algerian medicinal plants extracts containing phenolic compounds. Food Chem 97:654–60.
C:
Fo
od
Chemistr
y
Folch J, Lees M, Stanley GHS. 1957. A simple method for the isolation and purification of total lipids from animal tissues. J Biol Chem 226:497–509.
Gal S, Pinchuk I, Lichtenberg D. 2003. Peroxidation of liposomal palmitoyllinoleoyl-phosphatidylcholine (PLPC), effects of surface charge on the oxidizability and on the potency of antioxidants. Chem Phys Lipids 126:95–110.
Gulcin I. 2006. Antioxidant activity of caffeic acid (3, 4-dihydroxycinnamic acid). Toxicology 217:213–20.
Gutierrez ME, Garcıa AF, De Madariaga MA, Sagrista ML, Casado FJ, Mora M. 2003. Interac-tion of tocopherols and phenolic compounds with membrane lipid components: evaluaInterac-tion of their antioxidant activity in a liposomal model system. Life Sci 72:2337–60.
Halliwell B, Rafter J, Jenner A. 2005. Health promotion by flavonoids, tocopherols, tocotrienols, and other phenols: direct or indirect effects? Antioxidant or not?. Am J Clin Nutr 81:268S– 76S.
Heine RJ, Dekker JM. 2002. Beyond postprandial hyperglycaemia: metabolic factors associated with cardiovascular disease. Diabetologia 45(4):461–75.
Hildebrand A, Beyer, K, Neubert R, Garidel P, Blume A. 2004. Solubilization of negatively charged DPPC/DPPG liposomes by bile salts. J Colloid Interface Sci 279:559–571. Hopia A, Heinonen M. 1999. Antioxidant activity of flavonol aglycones and their glycosides in
methyl linoleate. J Am Oil Chemists’ Soc 76(1):139–44.
Ishimoto H, Tai A, Yoshimura, M, Amakura Y, Yoshida T, Hatano T, Ito H. 2012. Antioxidative properties of functional polyphenols and their metabolites assessed by an ORAC assay. Biosci Biotechnol Biochem 76(2):395–9.
Kenmogne-Domguia H, Moisan S, Viau M, Genot C, Meynier A. 2014. The initial character-istics of marine oil emulsions and the composition of the media inflect lipid oxidation during in vitro gastrointestinal digestion. Food Chem 152:146–54.
Laguerre M, Lecomte J, Villeneuve P. 2007. Evaluation of the ability of antioxidants to coun-teract lipid oxidation: existing methods, new trends and challenges. Prog Lipid Res 46:244– 82.
Liazid A, Palma M, Brigui J, Barroso CG. 2007. Investigation on phenolic compounds stability during microwave-assisted extraction. J Chromatogr A 1140:29–34.
Lorrain B, Dangles O, Genot C, Dufour C. 2010. Chemical modeling of heme-induced lipid oxidation in gastric conditions and inhibition by dietary polyphenols. J Agric Food Chem 58:676–83.
Meyer AS Heinonen M, Frankel EN. 1998. Antioxidant interactions of catechin, cyanidin, caffeic acid, quercetin, and ellagic acid on human LDL oxidation. Food Chem 61(1/2):71–75.
Mond´e A, Carbonneau MA, Michel F, Lauret C, Diabate S, Konan E, Sess D, Cristol JP. 2011. Potential health implication of in vitro human low-density lipoprotein-vitamin E oxidation modulation by polyphenols derived from C ˆote d’Ivoire’s oil palm species. J Agric Food Chem 59(17):9166–71.
Oueslati S, Trabelsi N, Boulaaba M, Legault J, Abdelly C, Ksouri R. 2012. Evaluation of antiox-idant activities of the edible and medicinal Suaeda species and related phenolic compounds. Ind Crops Products 36:513–8.
Peyrat-Maillard MN, Cuvelier ME, Berset C. 2003. Antioxidant activity of phenolic compounds
in 2, 2-azobis (2-amidinopropane) dihydrochloride (AAPH)-induced oxidation: synergistic
and antagonistic effects. J Am Oil Chemists’ Soc 80:1007–12.
Pinchuk I, Lichtenberg D. 2002. The mechanism of action of antioxidants against lipoprotein peroxidation, evaluation based on kinetic experiments. Prog Lipid Res 41:279–314. Proestos C, Komaitis M. 2008. Application of microwave-assisted extraction to the fast extraction
of plant phenolic compounds. Food Sci Technol 41:652–9.
Quettier-Deleu C, Gressier B, Vasseur J, Dine T, Brunet C, Luyckx M, Cazin M, Cazin J C, Bailleul F, Trotin F. 2000. Phenolic compounds and antioxidant activities of buckwheat (Fagopyrum esculentum Moench) hulls and flour. J Ethno-pharmacol 72:35–42.
Re R, Pellegrini N, Proteggente A, Pannala A, Yang M, Rice-Evans C. 1999. Antioxidant activity applying an improved ABTS radical cation decolorization assay. Free Radical Biol Med 26:1231–7.
Romani A, Pinelli P, Mulinacci N, Vincieri FF, Tattini M. 1999. Identification and quantitation of polyphenols in leaves of Myrtus communis. Chromatographia 49(1/2):17–20.
Singh H, Ye A, Horne D. 2009. Structuring food emulsions in the gastrointestinal tract to modify lipid digestion. Prog Lipid Res 48:92–100.
Vaya J, Mahmood S, Goldblum A, Aviram M, Volkovac N, Shaalana A, Musaa R, Tamira S. 2003. Inhibition of LDL oxidation by flavonoids in relation to their structure and calculated enthalpy. Phytochemistry 62:89–99.
Wannes WA, Mhamdi B, Sriti J, Jemia MB, Ouchikh O, Hamdaoui G, Kchouk ME, Marzouk B. 2010. Antioxidant activities of the essential oils and methanol extracts from myrtle (Myrtus communis var. italica L.) leaf stem and flower. Food Chem Toxicol 48:1362–70.
Yin J, Becker EM, Andersen ML, Skibsted LH. 2012. Green tea extract as food antioxidant.
Synergism and antagonism withα-tocopherol in vegetable oils and their colloidal systems.
